The Zurich Cohort Study is a prospective longitudinal study of a cohort of young adults from Zurich, Switzerland, who were followed from the ages of 20-40 in order to estimate the prevalence, incidence, impairment, treatment patterns and stability of the major mental disorders. This study assessed a wide range of psychiatric syndromes as well as somatic syndromes including mild manifestations under the conventional diagnostic thresholds (Angst et al 1984). The Zurich cohort study is comprised of a cohort of 4,547 subjects (m=2201; f = 2346) representative of the canton of Zurich in Switzerland, who were screened in 1978 with the Symptom Checklist 90-R (Derogatis 1988). In order to enrich the probability of psychiatric syndromes, a sub-sample of 591 subjects (291 males, 299 females) was selected for interview, with two thirds consisting of high scorers (defined by the 85th percentile or more of the SCL-90) and a random sample of those with scores below the 85th percentile. There have been a total of 6 waves of interviews covering the age period of 20-40 years. Those who had dropped out did not differ significantly in their baseline measures in terms of demographic characteristics, or risk group at study entry. A direct interview, the Structured Psychopathological Interview and Rating of the Social Consequences for Epidemiology (SPIKE), was administered by psychiatric residents and clinical psychologists with extensive clinical training in the subjects' homes. This interview schedule assesses a number of somatic syndromes, including headache, gastrointestinal, cardiovascular, and respiratory syndromes, as well as psychological syndromes, including depression, anxiety, phobia, obsessive-compulsive disorder, and substance abuse. Rather than adhering to a single diagnostic classification system, the entire spectrum of symptoms and correlates were assessed. Other measures included a personal and family history of all syndromes, dimensional measures of impairment and distress, personality traits, and other relevant constructs. Comorbid anxiety and depression tended to be far more persistent than either syndrome alone. Individuals with anxiety states alone tended to develop either depression alone or comorbid anxiety and depression as they progressed through adulthood. In contrast, depression alone and that which was comorbid with anxiety tended to be far more stable than anxiety alone over time.These findings have important implications for classification and treatment of affective disorders. The greater stability of comorbid anxiety and depression than either disorder alone illustrates the importance of further investigation of comorbid states compared to non-comorbid states in etiologic and treatment research. The persistence of subthreshold level depression and anxiety from early- to mid- adulthood suggests the importance of characterizing the continuum of expression of depression and anxiety rather than adhering to strict diagnostic thresholds. This study is an ongoing collaboration between the Principal Investigator and Professor Jules Angst in Zurich, Switzerland. The resources of our section have been devoted to the development of statistical models for longitudinal data on the evolution of mental disorders. During the past year, we have applied new approaches to investigate the longitudinal symmetry of anxiety and depression across 15 years of follow-up, and the role of bipolar disorder and social phobia in facilitating transitions of alcohol use to alcohol dependence. These papers will be completed during the next several months. We will then focus on the evolution of migraine symptoms and their association with anxiety and mood symptoms and disorders across early to mid adulthood.